the original thirteen NSFNET regionals

Tony Hain tonyhain at
Mon Jun 9 18:17:18 UTC 1997

Actually to complete the story, MFEnet preceded HEPnet & SPAN as DEC was
still developing the DECnet protocol at the time. When Sid put together
the proposal for SDSC, it was a clone of the operating and successful
MFE Computer Center (down to the floor plans of the building). The CTSS
operating system for the Cray at that time included the MFEnet protocol.

In that same time frame the MFEnet itself was running out of address
space and worked through a transition to TCP by running MFEnet over IP. 

	From:  Hans-Werner Braun [SMTP:hwb at]
	Sent:  Monday, June 09, 1997 10:19 AM
	To:  Kent W. England
	Cc:  nanog at
	Subject:  Re: the original thirteen NSFNET regionals

	> It's also worth noting that DECNET was a leading contender for
	> Sid Karin, director of SDSC, tells the story that he was
leaning toward
	> DECNET until he met a taciturn Teutonic network engineer from
Ann Arbor who
	> convinced Sid among others that a TCP/IP backbone would work.

	You actually do not quite have your story right. Sid used MFENET
	(Magnetic Fusion Energy) protocols for the SDSCnet (SDSC
	satellite network initially. I was also the networking person at
	University of Michigan responsible for the UMichigan-SDSCnet
	(with MFENET protocols) at that time (besides other
	It did work, and SDSC continued to use it despite the NSFNET
using IP.
	Over time SDSC phased MFENET out and IP in, but besides the
	of the taciturn Teutonic network engineer from Ann Arbor for the
	I don't think he had much to do with the phaseout of the MFENET
	at SDSC. Sid still claims that he is proud of both his MFENET
	as well as the decision to phase it out, as time moved on.

More information about the NANOG mailing list